Schizophrenia is a neurodevelopmental disorder in which cognitive control is impaired. This impairment is thought to be due, at least in part, to altered cortical GABA neurotransmission in parvalbumin (PV) expressing chandelier cells (ChCs). The axons of ChCs exclusively target the axon initial segment (AIS) of pyramidal neurons (PYR), the site of action potential generation, and thus ChCs can powerfully influence PYR cell output and prefrontal cortex (PFC) activity. The axon of a given ChC diverges to innervate hundreds of neighboring PYR cells, and the axons of multiple ChCs may converge onto a single PYR AIS, forming a distinctive, vertically-oriented array of terminals termed a cartridge. In PFC of subjects with schizophrenia, the density of cartridges detectable by GABA membrane transporter-1 (GAT1) immunoreactivity is 40% lower. This finding could reflect a disease process that severely affects a subpopulation of ChCs such that these ChCs furnish fewer axon terminals. As a result, some PYR AIS targeted by these ChCs do not receive a sufficient number of ChC terminals to form a recognizable cartridge. This interpretation is supported by findings showing that 1) in schizophrenia, ~50% of PFC PV neurons lack detectable levels of GAD67 mRNA; 2) in primate PFC, ChC terminals contain GAD67 protein, but low to no GAD65; and 3) in experimental systems depletion of GAD67 results in loss of terminals. Thus, we hypothesize that markedly lower GAD67 expression in a subset of ChCs results in fewer terminals innervating some PYR AIS such that some of those AIS are no longer associated with a recognizable presynaptic ChC cartridge. Understanding the pathogenesis of ChC abnormalities in schizophrenia requires knowledge of how ChC to PYR AIS inputs are refined during postnatal development. Preliminary data in monkey PFC show that the number of ChC inputs to PYR AIS is significantly greater in childhood than in adulthood. These findings suggest that the previously reported reduction in GAT1- immunoreactive cartridge density that occurs during adolescence in normal monkey PFC results from a decrease, or pruning, of the number of ChC terminals that innervate certain AIS. We hypothesize that ChC terminal inputs to PYR AIS are pruned during adolescence in monkey PFC, and that this reduction explains the prior observations of developmental changes in cartridge density. A finding that ChC terminal loss occurs during adolescence in monkeys would support the interpretation that our predicted results of fewer terminals per AIS in schizophrenia reflects excessive terminal pruning due to insufficient levels of GAD67 expression in ChC terminals during or before pruning of ChC to AIS connections. To test our hypotheses we use multi-label fluorescence confocal microscopy to assess the number of ChC terminals per AIS and relative mean GAD67 protein levels in ChC terminals in schizophrenia and across non-human primate development. Knowing the nature of the pathology at the AIS in schizophrenia and the developmental trajectory of reduced AIS innervation should provide insight into when the pathology at the AIS arises developmentally.